License plate location algorithm based on edge detection and morphology

Abstract

Vehicle license plate recognition has complex characteristics due to diverse effects as the fog, rain, shadows, irregular illumination conditions, partial occlusion, variable distances, cars' velocity, scene's angle and others. License-plate location in sensor images plays an important role in vehicle identification for automated transport systems. For the extraction of positioning in license plate recognition system, traditional Roberts operator and morphological edge detection algorithm exist the interference of image noise and low edge resolution. This paper proposes a new algorithm for vehicle license plate location based on edge detection and mathematical morphology. Roberts operator added two direction template can compensate the breaks of the edge and make it fully connected together. Then detect edge by selecting the appropriate threshold for Roberts operator and utilize morphological analysis to obtain candidate regions of license plate, analyze candidate regions by connected component, finally position plate region. The experimental results show that the algorithm can well suppress the impact of noise on the identification of the target area in the image, but also shorten the time of license plate location, valid license plate positioning.